At most one email a week. Unfollow anytime with one click.
Transcript
The full narration of this video. The current sentence is highlighted while the video plays; click any sentence to jump there.
Show transcript
Introduction
Hi, and welcome back. Once your first conversation goes well, two questions usually come up: which model should I use, and do I need a paid plan? In this episode we get to know Claude's model tiers, choose a model for each task, understand usage limits, and then compare plan types in a sensible way.
Step 1: What is a model?
When we type in Claude, what works behind the scenes is a model. Think of a model as the brain that processes our request and puts together the answer. Anthropic doesn't make just one model. There are several models in one family, each with different speeds and abilities, like a team whose members have different skills. The good news: you don't need to understand how models work technically. All you need to know is when to use a fast model and when to use the smartest one. To make it easy, picture three team members. First, the senior consultant: the smartest, great for complex problems, but takes longer to think and costs more. Second, the reliable staff member: balanced and quick. Most daily work, like writing emails or summarizing meeting notes, can go here. Third, the lightning assistant: the fastest and lightest, good for simple repetitive tasks, such as tidying a list or fixing spelling. So far, Claude's model family has been known by names like Opus, Sonnet, and Haiku. But names and versions keep getting updated, so check the latest list in the model picker and Anthropic's official documentation.
Step 2: Matching models to tasks
Let's compare the three tiers. In terms of speed, the smartest model is usually the slowest, while the fastest model answers almost instantly. In terms of ability, the smartest model excels at deep reasoning. The balanced model is already strong for most tasks, and the fastest model is enough for light tasks. One thing people often forget: quota use. Smarter models generally use up your usage allowance faster. This matters for free plan users. Example tasks: contract analysis or strategy for the smartest model, emails and captions for the balanced model, and tidying lists for the fastest model. The practical rule: make the balanced model your default. Move up to the smartest one only when the result isn't good enough or the task really is heavy. For example, Dewi from Maju Bersama Co. needs to review a draft supplier contract. This task needs careful reasoning, so she picks the smartest model in the model picker. Claude finds three clauses that could hurt the company, along with suggested changes. The answer takes longer to arrive, but for important decisions, accuracy is worth more than speed. On the other hand, when Dewi just wants to tidy a stock list, she picks the fastest model. The result is instant, and no quota is wasted. For serious legal matters, Claude's results are still discussed with an expert. The best way to learn the difference is to try it yourself. Send the same task to two models, then compare speed and quality. You'll quickly get a feel for when to move up a tier.
Step 3: Understanding usage limits
Now about usage limits. Dewi has been working in one long chat for a while, pasting twelve branch reports, and then asks for a ranking table. A notice appears saying the usage limit has been reached. Don't panic. This is normal, and Claude usually tells you when you can continue. The size of the limit differs from plan to plan and can change at any time. That's why we don't mention numbers; check Anthropic's official pages for the latest figures. While she waits, Dewi doesn't have to stop working. She can prepare her next data, or use a lighter model if her plan offers that option. There are four ways to save your allowance. First, open a new chat for a new topic. A very long chat gets reread every time you send a message. Second, choose the model that fits the task. Tidying a list doesn't need the smartest model. Third, attach only the parts you need. If you only need chapter three, don't attach the whole hundred-page report. Fourth, combine small requests into one clear message. One message with three requests uses less than three separate messages. These four habits do more than save your allowance. Claude's answers also tend to be more focused, because it isn't distracted by old conversations that are no longer relevant.
Step 4: Getting to know the plan types
Now about plans. In general there are three groups. First, the free plan, good for trying things out and light use. Every example in the early sessions of this class can be practiced on the free plan. Second, paid individual plans. The usage limits are bigger, and they usually unlock extra features, such as access to certain models or more complete work features. When this video was made, there was more than one individual plan tier. Third, plans for teams and enterprises. Beyond usage, there's member management, central billing, and stricter data controls. They suit offices like Maju Bersama Co. Plan names, prices, quotas, and features change often. So before you pay, always open Anthropic's official pricing page and read the details. So when should you subscribe? The first sign: you often hit the usage limit during work hours, and your work gets delayed. Second, Claude is already part of your core work every day, not just something you're trying out. Third, you need a feature that's only available on paid plans. The feature list for each plan is on the official page too. Fourth, and this is the most important, the time you save is worth far more than the cost of the subscription. On the other hand, if you've only tried it for two days and don't know your needs yet, there's no rush. Use the free plan first while you track your usage.
Step 5: Deciding with usage data
The most sensible way to decide is to track your usage. Laras makes a simple table. On Monday, she wrote a promo caption with the balanced model and saved twenty minutes. On Tuesday, shipping replies with the fastest model, saving fifteen minutes. On Wednesday, she analyzed hundreds of customer reviews with the smartest model. She saved forty-five minutes, but hit the usage limit along the way. On Thursday, a promo schedule with the balanced model, saving twenty-five minutes. The total is one hundred five minutes in a week, with the limit hit once. The formula at the top adds up the minutes column automatically. Laras's conclusion: for now, the free plan is still enough. If she hits the limit more often as her business grows, then she'll compare paid plans on the official page. Let's sum up the method. First, start with the free plan and use it for real tasks, not just trying things out. Second, track your usage for a week: the task, the model, the minutes saved, and how many times you hit the limit. Third, open Anthropic's official pricing page to see the latest options and features. Fourth, decide, then review it every few months, because both your needs and the plan options can change. This way, you pay because you actually need it, not because everyone else does. And the money you save can go to other needs in your business.
Done!
We've learned the three model tiers, matched them to tasks, understood usage limits and how to save them, and chosen a plan based on real usage notes. In the next episode we cover privacy, limitations, and how to use Claude safely, so your business and customer data stay protected. See you there!
Practice exercises
Try these on your own after watching to lock in what you learned.
Send the same task to two different models, then compare their speed and quality
Track your Claude usage for one week in a simple table
Read Anthropic's official pricing page and write down the plan that fits you best, with your reasons
Ask the class material
Type a question or keyword; we find the part of this episode that explains it and jump the video there.
My notes
Notes are saved with the current video time and are only visible to you.
Questions & answers
Stuck on a step? Ask here; the instructor and other learners can answer.
No questions yet.
Your instructor
Cerdas Inovasi Digital Team
Software developers, BIM & 3D engineers, and AI practitioners
53 video classes1,286 episodes
We build software, BIM models, games and AI automations for clients every day, and turn that hands-on work into step-by-step classes: every step is shown on screen and comes with a PDF guide you can keep.
